Predictive Analysis of Tool Wear Based on Vibration Signals in Turning

Machining processes on conventional lathes are often influenced by vibration phenomena, which can deteriorate surface finish, compromise dimensional accuracy, accelerate tool wear, and potentially induce chatter instability. Therefore, early detection of excessive vibration is critical for maintaining machining quality and preventing mechanical damage. The present study, titled “Sensor-Based Vibration Detection in Lathe Operations,” details the design and implementation of a real-time vibration monitoring system utilizing embedded sensors and signal processing techniques. The proposed system integrates a vibration sensor—either piezoelectric or MEMS-based accelerometer—mounted on the lathe structure proximate to the cutting zone to capture dynamic vibration signals generated during turning operations. The acquired analog signals undergo amplification and filtering stages before being digitized by a microcontroller-based data acquisition unit. The controller processes the digitized vibration signals to evaluate amplitude levels and detect abnormal operating conditions by comparing them against predefined threshold criteria. Upon exceeding safe vibration limits, the system triggers visual and/or audible alerts to notify the operator. Experimental validation conducted under varying spindle speeds, feed rates, and cutting depths demonstrates the system’s effectiveness in detecting chatter onset and irregular vibration patterns. This solution provides a cost-efficient and practical approach to condition monitoring, particularly suitable for small- and medium-scale workshops. By enabling timely fault detection and preventive intervention, the system enhances machining performance, reduces maintenance costs, and improves operational safety. Additionally, the research establishes a foundation for future integration with smart manufacturing and IoT-based predictive maintenance frameworks.